Petroleum refineries for SO2, NOx and TSP - CITEPA · Petroleum refineries for SO2, NOx and TSP...
Transcript of Petroleum refineries for SO2, NOx and TSP - CITEPA · Petroleum refineries for SO2, NOx and TSP...
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Petroleum refineries for SO2, NOx and TSP Index 0 INTRODUCTION ............................................................................................................2
1 DATA CURRENTLY USED IN THE RAINS MODEL ..............................................4
2 COMBUSTION IN REFINERY ...................................................................................10
3 GAS TURBINES IN REFINERY .................................................................................27
4 STATIONARY ENGINES IN REFINERY .................................................................28
5 FLUID CATALYTIC CRACKING UNIT...................................................................29
6 SULPHUR RECOVERY PLANTS ..............................................................................43
REFERENCES .......................................................................................................................48
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
0 Introduction 0.1 General information The purpose of refining is to convert natural raw materials such as crude oil into useful saleable products. Crude oil and natural gas are naturally occurring hydrocarbons found in many areas of the world in varying quantities and compositions. In refineries, they are separated and transformed into different products :
� Fuels for cars, trucks, airplanes, ships and other forms of transport; � Combustion fuels for the generation of heat and power for industry and households; � Raw materials for the petrochemical and chemical industries; � Specialty products such as lubricating oils, paraffins/waxes and bitumen; � Energy as a by-product in the form of heat (steam) and power (electricity).[1]
In order to manufacture these products, the raw materials are processed in a number of different refining facilities. The combination of these processing units to convert crude oil and natural gas into products, including its auxiliary units and facilities, is called a refinery [1]. Figure 1 shows the capacity of refinery in 25 Countries in Europe [8, 9]. Italy, Germany, France and UK have the largest capacities which represent more than 50% of the total capacity in Europe.
Capacities in Mt/Year
0
20
40
60
80
100
120
140
Italy
Ger
man
y
Fran
ce UK
Spai
n
Net
herla
nds
Belg
ium
Rom
ania
Swed
en
Gre
ece
Pola
nd
Nor
way
Portu
gal
Cro
atia
Hun
gary
Aust
ria
Finl
and
Den
mar
k
Bulg
ary
Switz
erla
ndSl
ovak
ia
Irela
ndC
zech
Rep
.
Figure 1: Refinery capacity in some European countries (for the year 2000)
The complexity of refineries has increased in the European Union with the installation of additional conversion units (thermal-, catalytic- and hydrocrackers). Now it must also be recognized that no two sites are the same, and the cost of installing particular facilities on one site may be very different from that for the same facilities on a different site – particularly if it is located in a different country.[2]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
0.2 Pollutants The main air emissions from a refinery are SO2, NOx and particulates, plus VOC which are not considered here. CONCAWE [2] made a study on SO2 released into the atmosphere from 70 refineries. In this report the relevant sources of SO2 are mentioned. As can be seen from the next table the largest source of SO2 (some 60 %) are fuel fired furnaces / boilers.
Table 0.1: sources of SO2 emissions in refinery [3]
Percentage of refinery SO2 emissions (%) Fuel fired in furnaces /boilers 59.4FCC units 13.5Sulphur Recovery Units 10.7Flares 5.0Miscellaneous 11.4Total 100 Two main sources of particulates exist in refineries :
1. Process heaters and boilers (burning oil) 2. Fluid catalytic Cracking Units (FCCU) and more specifically the catalyst regenerators
of such units. In general, fuel firing in furnaces and boilers represents the most important source of pollutants in a refinery.
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
1 Data currently used in the RAINS model 1.1 Sectoral aggregation of emission sources for the refinery sector
Table 1.1: RAINS sector of the SO2/NOx modules for stationary sources and their relation to the main activity groups of the CORINAIR inventory
RAINS sector Primary
Secondary
CORINAIR SNAP97 code
Fuel production and conversion (other than power plants) (CON)
- Combustion (CON_COMB) - Losses (CON_LOSS)
0103, 0104, 0105, 05
Industry (IN)
- Combustion in boilers, gas turbines and stationary engines (IN_BO) - Other combustion (IN_OC) - Process emissions (IN_PR)
0301 03 exc. 03011
04 Source: Sulfur emissions, abatement technologies and related costs for Europe in the RAINS model database. IIASA. June 1998.
Table 1.2: RAINS sector of the PM modules for refinery sector and their relation to the main activity groups of the CORINAIR inventory
RAINS sector RAINS code SNAP sector Fuel combustion in industrial boilers Combustion in boilers IN_BO
Combustion in boilers, grate combustion IN_BO1 Combustion in boilers, fluidized bed combustion IN_BO2 Combustion in boilers, pulverized fuel
b iIN_BO3
010301-03, 010501-03,
0301
Other combustion IN_OC Other combustion, grate combustion IN_OC1 Other combustion, fluidized bed combustion IN_OC2 Other combustion, pulverized fuel combustion IN_OC3
010304-06, 010504-06, 0302, 0303
Other industrial processes Petroleum refining PR_REF
Source: Modelling particulate emissions in Europe. A framework to estimate reduction potential and control costs. IIASA. 2002. 1.2 NOx abatement techniques used in the RAINS model The following section presents brief characteristics of the emission control technologies available for stationary sources. RAINS contains the following NOx control options for boilers and furnaces:
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
- Combustion modification (CM): Air staged Low NOx Burner (LNB), flue gas recirculation LNB, fuel staged LNB, fuel injection or reburning, fluidized bed combustion. - Selective catalytic reduction (SCR) - Selective non-catalytic reduction (SNCR) - Combined measures (combustion modification and SCR or SNCR)
Table 1.3: Main groups of NOx emission control technologies for stationary sources considered in RAINS
RAINS Sector/Technology Technology abbreviation
Removal efficiency
% Industrial boilers (IN_BO) and furnaces (IN_OC): CM - Solid Fuels CM - Oil&Gas CM+SCR Solid Fuels CM+SCR Oil &Gas CM+ Selective non-catalytic reduction (SNCR) Solid Fuels CM+SNCR Oil &Gas
ISFCM IOGCM ISFCSC IOGCSC ISFCSN IOGCSN
50 50 80 80 70 70
Process emissions: Stage 1 control Stage 2 control Stage 3 control
PRNOX1 PRNOX2 PRNOX3
40 60 80
Source: Nitrogen oxides emissions, abatement technologies and related costs for Europe in the RAINS model database. IIASA. October 1998. Cost evaluation methodology The model uses investment functions where these cost components are aggregated into one function. The shape of the function is described by its coefficients cif
and civ. The coefficients ci are given separately for three capacity classes: less than 20 MWth, from 20 to 300 MWth and above 300 MWth. When existing plant is retrofitted with add-on controls (SCR, SNCR) investments are multiplied by a retrofit cost factor r.
catcatv
fv
f cirbscici
bsciciI *)1(*)()( 2
21
1 λ+++++=
where: ci1
f, ci1v, ci2f, ci2
v – coefficients of investment function; ci1 have non-zero values only for combinations of technologies (e.g., CM plus SCR) bs – boiler size λcat
catalyst volume cicat unit cost of catalysts r retrofit cost factor Table 1.4: Coefficients of the investment function for add-on technologies and combined measures used in boilers and furnaces.
Technology abbreviation ci1
f, ci1v ci2f ci2
v Capacity range MWth
6.30 0 19.60 0 < 20 ISFCSC 5.18 22.50 14.60 102 20-300
52.33 876.50 5.10 2950 > 300
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
5.67 0 14.63 0 < 20 IOGCSC 4.66 20.25 11.25 68.85 20-300
2.10 788.85 4.73 1991.25 > 300 …
Source: Nitrogen oxides emissions, abatement technologies and related costs for Europe in the RAINS model database. IIASA. October 1998. Table 1.5: Other technology-specific parameters for add-on control technologies (secondary and combined measures) Parameter Unit Value Retrofit coefficient r %/100 0.5 Fixed O+M cost f %/100/yr 0.06 Catalyst cost cicat kECU/m3 10 Electricity demand λe - coal boilers GWh/PJ fuel input 0.36 - oil and gas boilers 0.30 Catalyst volume λcat Brown coal boilers Hard coal, dry bottom boilers Hard coal, wet bottom boilers Oil and gas boilers
m3/MWth
Sorbent demand λs, technology PBCSCR, PHCSCR, POGSCR PBCCSC, POGCSC PHCCSC, ISFCSC, IOGSCS ISFSCN, IOGCSN
t/t NOx
0.390 0.117 0.173 0.390
Source: Nitrogen oxides emissions, abatement technologies and related costs for Europe in the RAINS model database. IIASA. October 1998. 1.3 SOx abatement techniques used in the RAINS model
Table 1.6: Main groups of SO2 emission control technologies considered in RAINS
Technology name
RAINS abbreviation
Removal efficiency
% Use of low sulfur fuels (coal, and heavy fuel oil)
(·)
Limestone injection Industry
LINJ
50
Industry, Wet FGD (flue gas desulfurization) Power plants, Wet FGD, already retrofitted Power plants, Wet FGD High efficiency FGD
IWFGD PRWFGD PWFGD RFGD
85 90 95 98
Process emissions: Stage 1 control Stage 2 control Stage 3 control
SO2PR1 SO2PR2 SO2PR3
50 70 80
(·) The control efficiency depends on the initial sulfur content of the fuel to be replaced
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Source: Sulfur emissions, abatement technologies and related costs for Europe in the RAINS model database. IIASA. June 1998. Conventional Wet Flue Gas Desulfurization Processes Wet limestone flue gas desulfurization (WFGD) is the most commonly used flue gas desulfurization technique in Europe. In the early 1990s about 50.000 MWel of coal fired power plants were equipped with flue gas desulfurization, of which more than 80 percent were wet scrubbers (Vernon and Soud, 1990). This technology produces gypsum as a byproduct, which can be further used for a variety of industrial applications. WFGD processes have been installed in power plants, waste incineration plants and to some industrial heating plants. Early installations of WFGD processes were designed for sulfur removal efficiencies between 85 and 90 percent, while the latest installations reach up to 95 percent sulfur removal. High-efficiency Flue Gas Desulfurization In order to mark the upper end of available SO2 removal options, RAINS also considers high-efficiency processes while taking into account the increased costs of these options. There are several technical approaches to achieve sulfur removal rates up to 99 percent, e.g., specially designed wet FGD processes or the Wellman-Lord technology. RAINS uses the Wellman-Lord process to derive the typical economic and technical properties representative for such high-efficiency desulfurization techniques. This regenerative desulfurization method produces instead of waste material SO2 rich gas(about 97% SO2) that can be used as raw input to chemical industry to produce sulfuric acid or even elementary sulfur. Caustic soda (NaOH) is used as a sorbent. Spent absorber liquid is regenerated so that the losses of the sorbent are small. The desulfurization process is based on converting SO2 to sodium sulfates. Typical reduction efficiencies achieved have been more than 97 %. For add-on control options data distinguish technology-specific and country-specific parameters. The technology-specific parameters are common for all countries in Europe. The coefficients for calculating the investment functions are estimated separately for three capacity classes :
• < 20 MWth • 20-300 MWth • >300 MWth
Table 1.7: Technology-specific parameters for add-on control technologies
Parameter Unit Limestone
injection Wet FGD Advanced
FGD
Removal efficiency η Retrofit coefficient r Fixed O+M cost f Labor demand Electricity demand Sorbent demand Byproducts
% %/100 %/100/yr man-yr/GWth GWh/PJ fuel inp. t/tSO2 t/tSO2
50 0.3
0.04 10.8 0.5
4.68 7.8
95 0.3
0.04 10.8
1 1.56 2.6
98 0.3
0.04 25.2 2.2
0.01 0.5
Source: Sulfur emissions, abatement technologies and related costs for Europe in the RAINS model database. IIASA. June 1998.
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
1.4 PM abatement techniques used in the RAINS model
Table 1.8: Size-fraction specific removal efficiencies for abatement options used in RAINS for power plants and industry.
Removal efficiency Control technology RAINS code
> PM10 Coarse(1) Fine(2) Cyclone CYC, _CYC 90 % 70 % 30 % Wet scrubber WSCRB,
_WSCRB 99.9 % 99 % 96 %
Electrostatic precipitator, 1 field ESP1, _ESP1 97 % 95 % 93 % Electrostatic precipitator, 2 fields ESP2, _ESP2 99.9 % 99 % 96 % Electrostatic precipitator, 3 fields and more
ESP3P, _ESP3P 99.95 % 99.9 % 99 %
Wet electrostatic precipitator PR_WESP 99.95 % 99.9 % 99 % Fabric filters FF, _FF 99.98 % 99.9 % 99 % Regular maintenance, oil fired boilers GHIND 30 % 30 % 30 % (1): coarse particles: (> 2.5 and < 10 microns) (2): fine particles (< 2.5 microns) Source: Modelling particulate emissions in Europe. A framework to estimate reduction potential and control costs. IIASA. 2002. The petroleum refining industry converts crude oil into more than 2500 refined products, including liquid fuels (gasoline, diesel, residual oil), by-product fuels and feedstocks (e.g., asphalt, lubricants), and primary petrochemicals (e.g., ethylene, toluene, xylene). RAINS Sector: PR_REF IIASA decided to use the value from the Dutch inventory [14].
Table 1.9: Emission factors used in the RAINS model for refineries [kg/t crude oil].
Sector RAINS code PM2.5 Coarse PM10 >PM10 TSP Petroleum refining
PR_REF 0.096 0.024 0.120 0.002 0.122
Source: Modelling particulate emissions in Europe. A framework to estimate reduction potential and control costs. Page 69. IIASA. 2002. The RAINS model includes cyclones, bag filters and electrostatic precipitators as control options for refineries.
Activities in some countries for the RAINS sector PR_REF The baseline for the EU-15 of the energy pathway is the PRIMES model.
Table 1.10: Activity for some countries of the EU-15 (Mt)
Country 1990 1995 2000 2005 2010 Belgium 27.20� 25.63� 29.60� 30.51� 31.52�France 75.40� 79.45� 85.08� 99.62� 96.77�
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Italy 89.10� 79.52� 85.50� 75.91� 72.31�Germany New Länder 15.70� 17.42� 18.10� 17.80� 17.51�
Germany Old Länder 78.20� 86.91� 90.29� 88.80� 87.33�
Spain 54.60� 54.93� 61.53� 71.08� 72.98�United Kingdom 89.60� 86.63� 78.51� 90.09� 92.41�…
Others information such as Emission factors can be found in the Web PM module: http://www.iiasa.ac.at/~rains/cgi-bin/rains_pm
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
2 Combustion in refinery 2.1 General information SNAP CODE: 01 03 01 -> 01 03 03 + 01 03 06 NFR: 1b Sector activity unit: PJ fuel input SO2 NOx PM VOC NH3 X X X - -
This sector covers emissions from power plants in refineries producing steam and/or electricity . Combustion engines and gas turbines are not included. The combustion processes in refineries for the heating of petroleum products without direct contact between flame or flue gas and products are considered too. 2.2 Definition of reference installation/process Fired boilers and furnaces generate substantial SO2, NOx and particulate emissions, particulary when heavy fuel oil is used. Gas-fired boilers generate hardly any dust and low SO2 emissions, when the refinery gases are cleaned in amine scrubbers. NOx emissions are also much lower than those of oil-fired boilers. CONCAWE leaded a “Review of the Cost Effectiveness of NOx Control Measures on Combustion Units in European Refineries” (Input to UN-ECE EGTEI). This document gives data on range of sizes of combustion units in European refineries derived from survey data on some 100 combustion units.
There are some 100 Refineries in Europe and each of those Refineries will have several heaters/furnaces; something of the order of a 1000. CONCAWE does not have information on all those heaters but has enough data to do a representative sample, a selection of 50. Their
Variation of Total Heat Fired Across Surveyed Refineries
0
50
100
150
200
250
300
350
400
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49Stack Number
Tota
lFiri
ng(M
W)
Variation of Total Heat Fired Across Surveyed Refineries
0
50
100
150
200
250
300
350
400
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49Stack Number
Tota
lFiri
ng(M
W)
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
size varied from 7MW up to 386 MW. Here the range 30-150 MW was taken as representative. The heaters were geographically dispersed from Scandinavia to Italy. According these statistics, the expert group choose a capacity of 50 MWth.
Table 2.1: Reference installations
Reference Code Technique Fuel Capacity [MWth]
Life time [a]
01 Combustion unit Heavy fuel oil 50 30 02 Combustion unit Gas 50 30
2.3 Emission abatement techniques and costs This chapter considers the emission abatement techniques for combustion processes in a refinery, the cost of installing and operating them and their performance. In refineries, secondary measures are only applied to larger units, since new units are rarely built and the specific costs for retrofitting of smaller units are considerably higher than for large units. In some cases, retrofitting secondary measures is not technically feasible due to space constraints. The most representative measures considered in the EU-countries are mentioned below. 2.3.1 PM emissions No dust abatement techniques are installed in refinery boilers. Table 2.2: TSP emissions level for each reference installation
Description EF TSP [mg/Nm³]
mean value
EF TSP [g/GJ fuel input]
mean value Reference Installation 01
Uncontrolled 200-1000 56-280 Reference Installation 02
Uncontrolled 5 1.35 Remarks:
1. An average conversion factor (Fconv) between concentrations of pollutants (in mg/Nm3) and specific mass flows of pollutants (emission factor, in g per GJ fuel input) [4] Concentration of pollutant emitted (in mg/Nm3) x Fconv = Specific mass flow of pollutant emitted (in mg/GJ fuel input)
For liquid fuels: Fconv= 280 Nm3/GJ (3 % O2, dry)
For gaseous fuels: Fconv= 270 Nm3/GJ (3 % O2, dry)
2. To determine exactly the TSP Emission factor for liquid fuel firing, three parameters are needed :
- Edust: Total dust emissions in a given country (from emission inventory) - Econs gas: Gas consumption (GJ) - Econs fuel: Liquid fuel consumption (GJ)
The dust emissions caused by gas firing Edust GF is equal to the EF TSP for the reference installation 2 · Econs gas.
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Then the dust emissions caused by liquid fuel firing: Edust LF = Edust - Edust GF. Knowing the liquid fuel consumption, the EF for the reference installation 1 can be determined.
3. To abate the dust emission of heavy fuel firing boilers, a fuel switch with gas can be used. The application rate of this fuel switch is country-specific. 2.3.2 NOx emissions 2.3.2.1 Abatement measures
� Low NOx burner Low NOx burners have the aim of reducing peak flame temperature, reducing oxygen concentration in the primary combustion zone and reducing the residence time at high temperature, thereby decreasing thermally formed NOx.[2]
� SNCR For the SNCR technology, an additive (ammonia or urea) is injected into the combustion chamber of the waste incineration plant. The conversion of nitrogen oxides into nitrogen and water takes place at temperatures between 850 and 1,100°C without a catalyst. In order to achieve satisfactory performance of the SNCR technology, the required temperature window has to be respected. It is necessary that the injection nozzles are disposed at several locations in the combustion chamber to overcome the inhomogeneous composition of waste, as well as the resulting variations in the temperature profile within the combustion chamber.
� SCR
In SCR De-NOx systems, NOx contained in many kinds of exhaust gases is reduced by ammonia (NH3), urea [(NH2)2CO], etc. called "Ammonia like material" to nitrogen (N2) and water (H2O), on a catalyst. The most suitable reducing agent can be selected out of ammonia like materials based on economics, handling and safety criteria. SCR De-NOx system mainly consists of a reactor, reducing agent, injection system and catalyst. After injection and complete mixing of reducing agent with the gas at the inlet of reactor, the exhaust gas is led into a catalyst bed. NOx is converted into N2 and H2O on the catalyst surface. When reducing agent is NH3, chemical reactions are represented as follows : 4NO+4NH3+O2 -> 4N2+6H2O
6NO2 +8NH3 -> 7N2+12H2O
Table 2.3: NOx abatement measures for each reference installation
Description EF NOx [mg/Nm³]
mean value
EF NOx [g/GJ fuel input]
mean value
Abatement efficiency for
NOx [%]
Reference Installation 01 Uncontrolled 600 168 -
Low NOx Burner 420 101 30 SNCR(1) 170 48 60 SCR(1) 65 18 85
Reference Installation 02
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Uncontrolled 200 54 Low NOx Burner 100 27 50
SNCR(1) 40 11 60 SCR(1) 15 4 85
(1): after having installed the low NOx burner technology .
2.3.2.2 Costs For determining the different costs, information provided by CONCAWE and by ADEME have been used [1, 15].
� Low NOx burner
Reference Capital Costs for 28 MW Unit: 200-600 k€
Reference Operating Cost for 28 MW Unit: Zero
Cost vs Unit Size = Cost Ref · [MW/Mwref]^0.8
In the case of a reference installation with a capacity of 50MW,
Capital Costs for 50 MW liquid fuel firing Unit = 400·[50/28]^0.8
= 636 k€
Capital Costs for 50 MW gas firing Unit = 250·[50/28]^0.8
= 400 k€
The different investments costs proposed by ADEME, which are derived from French refineries [15] are detailed in the Excel sheet established. From these costs, the average investment is 450 k€ for fuel firing boiler and 300 k€ for gas firing boiler.
� SNCR
Reference Capital Costs for 28 MW Unit: 400-900 k€
Reference Operating Cost for 28 MW Unit: 25-70 k€/y
Cost vs Unit Size = Cost Ref · [MW/Mwref]^0.6
In the case of a reference installation with a capacity of 50MW,
Capital Costs for 50 MW liquid fuel firing Unit = 650·[50/28]^0.6
= 920 k€
Capital Costs for 50 MW gas firing Unit = 500·[50/28]^0.6
= 700 k€
The different investments costs proposed by ADEME, which are derived from French refineries [15] are detailed in the Excel sheet established. From these costs, the average investment is 920 k€ for fuel firing boiler and 700 k€ for gas firing boiler.
� SCR
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Reference Capital Costs for 28 MW Unit: 2.8-3.2 M€
Reference Operating Cost for 28 MW Unit: 150 k€/y
Cost vs Unit Size = Cost Ref · [MW/Mwref]^0.6
In the case of a reference installation with a capacity of 50MW,
Capital Costs for 50 MW liquid fuel firing Unit = 3·[50/28]^0.6
= 3.89 M€
Capital Costs for 50 MW gas firing Unit= 2.8·[50/28]^0.6
= 3.6 M€
The different investments costs proposed by ADEME, which are derived from French refineries [15] are detailed in the Excel sheet established. From these costs, the average investment is 2,000 k€ for fuel firing boiler and 1,900 k€ for gas firing boiler.
Table 2.4: Investments and Operating costs
Description Investment (k€) Fixed Operating costs (%/a)*
Variable Operating costs (k€/PJ/a)
Source (1) Source (15) Reference installation 1
Low NOx Burner 636 450 4 0 SNCR 920 920 4 32 SCR 4,150 2,000 4 55.7
Reference installation 2 Low NOx Burner 400 300 4 0
SNCR 700 700 4 15.6 SCR 3,850 1,900 4 44.2
*: The fixed Operating costs only depend on the capacity - or size - of the installation, i.e. on the investment, and they are expressed as a percentage of the plant investment [%/a] Parameters needed to calculate Variable Operating costs
� SNCR Electricity cost λe · ce / 10-3 [k€/PJ fuel input]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/PJ fuel input]
• ce: electricity price [€/kWh] λe = 5.56·104 kWh/PJ for a consumption of 10 kW [15] (see Excel sheet for more details) ce = 0.0569 €/kWh (value for France) Ammonia cost λs · cs · efunabated · η / 103 [k€/PJ fuel input]
• efunabated: unabated emission factor of pollutant [t pollutant/PJ fuel input] • λs: specific ammonia demand [t/t pollutant removed] • cs: ammonia price [€/t]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
• η: removal efficiency = (1 - efabated/efunabated)
λs = λm·λM/η
with: λm: NH3/ NOx (mol/mol) ratio for NOx emitted λM: NH3/ NOx (mol weight/mol weight) ratio λs = 1.5/0.6· (17/46) = 0.92 efunabated = 100.8 t NOx/PJ for fuel firing efunabated = 27 t NOx/PJ for gas firing λs = 0.92 tNH3/t NOx removed cs = 400 €/tNH3 (ammonia pur) η = 60 % Labour cost (λ l · cl) · 106 / (3,6 · pf) [k€/PJ fuel input]
• λ l: labour demand [person-year/MWth] or [man -year/GWth] • cl: labour cost/wages [k€/ man -year] • pf: plant factor [h/a]
The number of additional personnel required for the SNCR unit is taken here as 0.25
Thus, the annual personnel costs for the SNCR process are:
ACPERS = 0.25 � cl
Thus λ l = (0.25)/ Capacity
= 0.25/50
= 5�10-3 person-year/MWth
λ l = 5�10-3 person-year/MWth cl = 37,234 k€/ person –year (value for France) pf= 8,000h
Table 2.5: Parameters needed to calculate variable Operating costs for SNCR
efunabated
[g NOx/GJ] ηηηη λλλλs
[t/t NOx removed]
cs
[€/t] λλλλe
[kWh/PJ]ce
[€/kWh]
λλλλ l
[person-year/MWth]
cl [€/person-year]
pf [h/a]
Reference installation 1 SNCR 100.6 60 0.92 400 5.56·104 0.0569 0.005 37.234 8,000
Reference installation 2 SNCR 27 60 0.92 400 5.56·104 0.0569 0.005 37.234 8,000
� SCR
Electricity cost λe · ce / 10-3 [k€/PJ fuel input]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/PJ fuel input]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
• ce: electricity price [€/kWh] λe = 3.33·105 kWh/PJ for a consumption of 60 kW [15] (see Excel sheet for more details) ce = 0.0569 €/kWh (value for France) Ammonia cost λs · cs · efunabated · η / 103 [k€/PJ fuel input]
• efunabated: unabated emission factor of pollutant [t pollutant/PJ fuel input] • λs: specific ammonia demand [t/t pollutant removed] • cs: ammonia price [€/t] • η: removal efficiency = (1 - efabated/efunabated)
λs = λm·λM/η
with: λm: NH3/ NOx (mol/mol) ratio for NOx emitted λM: NH3/ NOx (mol weight/mol weight) ratio λs = 1.05/0.85· (17/46) = 0.46 efunabated = 100.8 t NOx/PJ for fuel firing efunabated = 27 t NOx/PJ for gas firing λs = 0.46 tNH3/t NOx removed cs = 400 €/tNH3 (ammonia pur) η = 85 % Labour cost [person-year/MWth]: (λ l · cl) · 106 / (3,6 · pf) [k€/PJ fuel input]
• λ l: labour demand [person-year/MWth] or [person -year/GWth] • cl: labour cost/wages [k€/ person -year] • pf: plant factor [h/a]
The number of additional personnel required for the SCR unit is taken here as 0.25
Thus, the annual personnel costs for the SCR process are:
ACPERS = 0.25 � cl
Thus λ l = (0.25)/ Capacity
= 0.25/50
= 5�10-3 person-year/MWth
λ l = 5�10-3 person-year/MWth cl = 37,234 k€/ person –year (value for France) pf= 8,000h Catalyst cost (λcat · cicat / ltcat) · (103 /3.6) [k€/PJ fuel input]
• λcat: catalyst volume [m3/MWth] • cicat: unit costs of catalysts [k€/m3] • ltcat: life time of catalyst [103 h]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
and: 610].[6,3
].[][h
MWPJ =
In our case, the gas volume flow is around: V = Fconv·(Capacity) = 275·(50·10-3·3600) = 48,600 Nm³/h Then according some statistics [15], the catalyst volume is around 7 m³. cicat = 15 k€/m3 λcat = 0,14 m3/MWth [15] (see Excel sheet for more details) ltcat = 5 years = 40 · 103 h pf= 8,000 h
Table 2.6: Parameters needed to calculate variable Operating costs for SCR
efunabated
[g NOx/GJ] ηηηη λλλλs
[t/t NOx removed]
cs
[€/t] λλλλe
[kWh/PJ] ce
[€/kWh]
λλλλ l
[person-year/MWth]
cl
[k€/ person -year]
pf [h/a]
λλλλcat
[m³/MWth] cicat
[k€/m3]ltcat
[103 h]
Reference installation 1 SCR 101 85 0.46 400 3.33·105 0.0569 0.005 37.234 8,000 0.14 15 40
Reference installation 2 SCR 27 85 0.46 400 3.33·105 0.0569 0.005 37.234 8,000 0.14 15 40
� Conclusion
In the petroleum industry as the SCR process is much more likely to be chosen than the SNCR process, the cost of secondary measures has been assessed taking into account the following shares : 99 % of SCR 01 % of SNCR. According to this repartition, the different costs of the NOx secondary measures are the following:
Table 2.7: Investments and Operating costs of the secondary measures
Description Lifetime Investment (k€) Fixed Operating costs
Variable Operating costs
(a) Source (1) Source (15) (%/a) (k€/PJ) Reference installation 1
None - - - - Secondary technology 10 4,120 2,000 4 55,7
Reference installation 2 None - - - -
Secondary technology 10 3,820 1,900 4 44,2
Table 2.8: Parameters needed to calculate variable Operating costs for secondary measure efunabated
[t NOx/t] ηηηη λλλλs
[t/t NOx removed]
cs
[€/t] λλλλe
[kWh/t] ce
[€/kWh]
λλλλ l
[person-year/t]
cl
[k€/person-year]
λλλλcat
[m³/t] cicat
[k€/m3]ltcat
[103 hrs]
Reference installation 1 1001 85 0.46 400 3.33·105 0.0569 0.005 37.234 0.14 15 8,000
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Reference installation 2 27 85 0.46 400 3.33·105 0.0569 0.005 37.234 0.14 15 8,000
2.3.2.3 Methodology to calculate the application rate To determine the application rate of the different abatement measures, the following methodology can be used. Input parameter :
- Econs gas: Gas consumption (GJ) - Econs fuel :Liquid fuel consumption (GJ) - ENOx: Emission of NOx in a country (t per year)
Considering the different emission factors determined in the last paragraph, it is possible to calculate an equivalent Emission level for a fuel mixture of gas and liquid fuel : Equivalent Emission level = [(NOx emission factor for Reference installation 1) x Econs gas + (NOx emission factor for Reference installation 2) x Econs fuel] / (Econs gas + Econs fuel)
Table 2.9: Emission level for fuel mixture
Description EF NOx [g/GJ fuel input] for Gaseous fuel
EF NOx [g/GJ fuel input]for Liquid fuel
Emission level [g/GJ fuel input] for fuel mixture
Uncontrolled 168 54 (168·Econs gas +54· Econs fuel)/ (Econs gas + Econs fuel)
Low NOx Burner 101 27 (101·Econs gas +27·Econs fuel)/ (Econs gas + Econs fuel)
Low NOx Burner + secondary measure
18 4 (18·Econs gas +4·Econs fuel)/ (Econs gas + Econs fuel)
Then, the sector situation may be defined by:
Fs a NOx = ENOx / (Econs gas + Econs fuel)
According to this result, it is possible to calculate the different application rates : FS1NOx fuel mix : Uncontrolled NOx emission level for fuel mixture FS2NOx fuel mix : NOx emission level implementing the DeNOx stage 1 technical option
(primary measures - PM) for fuel mixture FS3NOx fuel mix : NOx emission level implementing the DeNOx stage 2 technical option
(secondary measures - SM) for fuel mixture
� If FS1NOx > Fs a NOx > FS2NOx, it can be considered that some primary measure may still be implemented on a given percentage of the production capacity.
The virtual application rate of primary measures T1,NOx is obtained by:
T1,NOx = (Fs a NOx - FS1NOx fuel mix)/(FS2NOx fuel mix - FS1NOx fuel mix)
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
� If Fs a NOx < FS2NOx it may be considered that some secondary measures have already been implemented. In this case, it can be considered that the application rate concerning NOx primary measures is 100%.
The virtual application rate of secondary measures T2,NOx is obtained by:
T2,NOx = (Fs a NOx - FS2NOx fuel mix) / (FS3NOx fuel mix - FS2NOx fuel mix)
Table 2.10: Application rate and applicability for NOx abatement measures
Description Application rate in 2000
[%]
Application rate in 2005
[%]
Applicability [%]
Application rate in 2010
[%]
Applicability [%]
Application rate in 2015
[%]
Applicability [%]
Application rate in 2020
[%]
Applicability [%]
None Primary
technologies 100 100 100 100
Secondary technologies 100 100 100 100
2.3.3 SOx emissions 2.3.3.1 Abatement measures
� Fuel switch HF-GAS The aim of the fuel switch is to change to a fuel leading to a less pollutant emissions. In the case of refinery a fuel switch from heavy fuel oil to gas is normally realized, lowering emission of NOx, SO2 and particulate matter.
� Wet scrubber
Table 2.11: SOx abatement measures
Description EF SOx [mg/Nm³]
mean value
EF SOx [t/PJ fuel input]
mean value
Reference Installation 01 Uncontrolled 3,400* 950
Fuel switch HF-GAS 20 5.4 Scrubber (η=90 %) 340 95
Reference Installation 02 Reference level 20 5.4
*: depending of the characteristics of the fuel. In this case, it is heavy fuel oil with 2% S content (3 % O2, dry). 2.3.3.2 Costs Table 2.12: Investments and Operating costs
Description Investment (k€) Fixed Operating costs (%/a)
Variable Operating costs
(k€/PJ)
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Fuel switch HF-GAS 0 4 See table 2.13 Wet scrubber* 4,000 4 433
*This option is detailed in the following paragraph, although CONCAWE doesn’t validate this abatement option. Parameters needed to calculate Variable Operating costs
� Fuel switch HF-GAS [k€/PJ fuel input] SUM [(new consump/prod)i · (new price)i] – SUM [(old consump/prod)i · (old price)i]
• Energy consumption (+) • Energy production (-) • Consumption [MJ/GJth input] • Fuel price [€/GJ]
Table 2.13: Parameters needed to calculate variable Operating costs for fuel switch (new consump/prod)I
[MJ/GJth input] (new price)I
[€/GJ] (old consump/prod)I
[MJ/GJth input] (old price)I
[€/GJ]
� Wet scrubber
The different costs are the following: Limestone cost: λs � cs � efunabated � η / 103 [k€/PJ]
• efunabated: unabated emission factor of pollutant [t pollutant/PJ] • λs: specific limestone demand [ton/t pollutant removed] • cs: limestone price [€/t] • η: removal efficiency (= 1 - efabated/efunabated)
with: λs = λm � λM λm: Ca/S (mol/mol) ratio λM: CaCO3/SO2 (mol weight/mol weight) ratio
efunabated = 950 t SO2 /PJ η = 90 % λs = 1.59 t/tSO2 removed cs = 20 €/t (value for France)
Waste disposal cost λd � cd � efunabated � η / 103 [k€/PJ]
• efunabated: unabated emission factor of pollutant [t pollutant/PJ] • λd: demand for waste disposal [ton/ t pollutant removed] • cd: byproduct/waste disposal cost [€/ton] • η: removal efficiency (= 1 - efabated/efunabated)
efunabated = 950 t SO2 /PJ η = 90 %
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
λs = 2.6 t/tSO2 removed cs = €/t (value for France)
Labour cost λ l � cl [k€/PJ]
• λ l: labour demand [person-year/MWth] • cl: labour cost/wages [k€/person-year]
The number of additional personnel for the wet scrubber is taken here as 0.5 person-year.
Thus, the annual personnel costs are:
ACPERS = 0.5 � cl
Thus λ l = (0.5)/ Capacity
= 0.5/50
= 0.01 person-year/MWth
cl = 37,234 k€/ person-year (value for France) λ l = 0.01 person-year/t Electricity cost λe � ce / 10-3 [k€/PJ]
• λe: additional electricity demand (= new total consumption – old total consumption) [kWh/PJ]
• ce: electricity price [€/kWh] λe = 1.06�106 kWh/PJ for 190 kW ce = 0.0569 €/kWh (value for France) Table 2.14: Parameters needed to calculate variable Operating costs for Wet scrubber
efunabated
[t SO2/PJ] ηηηη λλλλs
[t/t SO2 removed]
cs
[€/t]
λλλλd
[t/t SO2 removed]
cd
[€/ton] λλλλ l
[person-year/PJ]
cl [k€/person
-year]
λλλλe
[kWh/t] ce
[€/kWh]
Variable Operating
costs (k€/PJ)
Wet scrubber 950 90 1.59 20 2.6
0.01 37,234 1.06�106 0.0569 433
2.3.3.3 Methodology to calculate the application rate To determine the application rate of the different abatement measure, the following methodology can be used. Input parameter :
- Econs fuel: Liquid fuel consumption (GJ) - ESOx: Emission of SOx in a country (t per year) for liquid fuel firing
Then, the sector situation may be defined by:
Fs a SOx = ESOx / Econs fuel
According to this result, it is possible to calculate the different application rates :
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
FS1SOx fuel : Uncontrolled SOx emission level for liquid fuel FS2SOx fuel : SOx emission level implementing the Wet scrubber
The virtual application rate of the wet scrubber T2,SOx is obtained by:
T2,SOx = (Fs a SOx - FS1SOx fuel )/(FS2SOx fuel - FS1SOx fuel)
Table 2.15: Application rate and applicability for SOx abatement measures
Description Application rate in 2000
[%]
Application rate in 2005
[%]
Applicability [%]
Application rate in 2010
[%]
Applicability [%]
Application rate in 2015
[%]
Applicability [%]
Application rate in 2020
[%]
Applicability [%]
None Fuel switch
HF-GAS 100 100 100 100
Wet scrubber 100 100 100 100
2.4 Regulatory constraints introduced by the LCP Drective
2.4.1 SO2 emissions
Concerning SO2 emissions from boilers, it is important to distinguish SNAP codes : - 01 03 01: Combustion plants >= 300 MW (boilers)
- 01 03 02: Combustion plants >= 50 and < 300 MW (boilers)
- 01 03 03: Combustion plants < 50 MW (boilers) The refinery sector uses the bubble concept which refers to air emissions of SO2. This concept is a regulatory tool applied in several EU countries. The bubble approach for emissions to air reflects “a virtual single stack” for the whole refinery. This concept is used for refineries because it is recognised that they meet some or all of their energy needs with a variety of gaseous and liquid fuels that are by-products of the various processes. [18] In the revisions of the LCP Directive, the SO2 emission limit values VEL for new and existing refineries are respectively 600 and 1000 mg/Nm³. With 2 input parameters, it is possible to calculate how many emissions a country does abate in order to comply with LCP Directive. Input Parameters :
- Econs fuel: Liquid fuel consumption (GJ) - ESOx: Emission of SOx in a country (t per year) for liquid fuel firing
Then, the sector situation may be defined by:
Fs a SOx = ESOx / Econs fuel
Taking into account the average conversion factor (Fconv) between concentrations of pollutants (in mg/Nm3) and specific mass flows of pollutants (emission factor, in g per GJ fuel input) for liquid fuels :
Fconv= 280 Nm3/GJ (3 % O2, dry)
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
The situation versus the LCP Directive then is :
VEL101
EE
S 9
lfue cons
SOx −⋅⋅=convF
[mg/Nm³]
If the result is positive, the country has to abate “S” mg/Nm³ to follow the LCP Directive. Otherwise the country has abated more than the Directive imposed. 2.4.2 NOx and Dust emission For NOx and Dust emissions, the LCP Directive contains a lot of constraints, depending on the state of the installation, the capacity and the fuel used. Examplary of Sweden The Swedish Parliament decided in 1990 to introduce a tax to be paid for emissions of nitrogen oxides from boilers with a usable energy production of at least 50 gigawatt hours (GWh) per year. The NOx charge is based on actual recorded emissions. The abatement cost was found to be between 3 and 84 SEK/ kg of NOx reduced, 331 to 9,270 €/t of NOx reduced The charge is imposed irrespective of the fuel used and is levelled at a rate of SEK 40 per kg of emitted NOx, 4412 €/t of emitted NOx. To avoid distorting the pattern of competition between those plants which are subject to the NOx charge and those that are not (and possibly create incentives to replace existing equipment with inefficient smaller boilers that are not subject to the charge), the system is designed in a way that all revenue except the cost of administration is returned to the participating plants, in proportion to their final production of usable energy. [19] The consequence of this charge in Sweden is that the average cost of measures to reduce emissions as a result of the charge on NOx was SEK 7.5 per kg of NOx reduced, about 900 €/t of NOx reduced (according to a 1996 study). [19]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Combustion in refinery Summary list of parameters and data
Parameter Annotatio
n Unit Type of data Current proposal
1 Gas consumption 2000, 2005, 2010, 2015 and 2020
Econs gas GJ Input -
2 Liquid fuel consumption 2000, 2005, 2010, 2015 and 2020
Econs fuel GJ Input -
3 SOx (as SO2) ESO2 Tonnes per year Input - 4 NOx (as NO2) ENOx Tonnes per year Input - 5 Dust EDust Tonnes per year Input - 6 Conversion factor between concentration
and specific mass flow for liquid fuel Fconv Lfuel Nm3/GJ Fixed by the
experts 280
7 Conversion factor between concentration and specific mass flow for gas
Fconv gas Nm3/GJ Fixed by the experts
270
8 Uncontrolled dust emission level for liquid fuel firing
FS Dust Lfuel mg/Nm³
g/GJ fuel input
Fixed by the experts
200-1000
56-280 9 Uncontrolled dust emission level for gas
firing FS Dust gas mg/Nm³
g/GJ fuel input
Fixed by the experts
5
1.35 10 Uncontrolled NOx emission level for
liquid fuel firing FS1NOx Lfuel mg/Nm³
g/GJ fuel input
Fixed by the experts
600
168 11 NOx emission level implementing the
DeNOx stage 1 technical option (primary measures - PM) for liquid fuel firing
FS2NOx Lfuel mg/Nm³
g/GJ fuel input Fixed by the
experts
420
101
12 NOx emission level implementing the secondary measure for liquid fuel firing
FS3NOx Lfuel mg/Nm³
g/GJ fuel input
Fixed by the experts
65
18 13 Cost of the DeNOx stage 1 technical
option (PM) per tonne of pollutant avoided
C NOx 1 Lfuel Euro / tonne NOx
abated
Evaluated by the experts
1,650 (1)
1,012 (2) 14 Cost of the DeNOx secondary measure
per tonne of pollutant avoided C NOx 2 Lfuel
Euro / tonne NOx abated
Evaluated by the experts
6,100 (1)
3,290 (2) 15 Uncontrolled NOx emission level for gas
firing FS1NOx gas mg/Nm³
g/GJ fuel input
Fixed by the experts
200
54 16 NOx emission level implementing the
DeNOx stage 1 technical option (primary measures - PM)
FS2NOx gas mg/Nm³
g/GJ fuel input Fixed by the
experts
100
27
17 NOx emission level implementing the secondary measure
FS3NOx gas mg/Nm³
g/GJ fuel input
Fixed by the experts
15
4 18 Cost of the DeNOx stage 1 technical
option (PM) per tonne of pollutant avoided
C NOx 1 gas Euro / tonne NOx
abated
Evaluated by the experts
1,940 (1)
1,260 (2) 19 Cost of the DeNOx secondary measure
per tonne of pollutant avoided C NOx 2 gas
Euro / tonne NOx abated
Evaluated by the experts
20,800 (1)
11,260 (2) 20 Uncontrolled SO2 emission level for
liquid fuel firing used for the economical assessment
FS1 SO2 Lfuel mg/Nm³
g/GJ fuel input
Fixed by the experts
3,400
950 21 SO2 emission level implementing the
switch from liquid to gaseous fuels FS2 SO2 Lfuel mg/Nm³
g/GJ fuel input
Fixed by the experts
20
5.4 22 Applicability rate for the fuel switch AR % Input - 23 SO2 emission level implementing the wet
scrubber technical option FS3 SO2 Lfuel mg/Nm³
g/GJ fuel input
Fixed by the experts
340
95 24 Cost of the stage 1 DeSO2 technical
option per tonne of pollutant avoided for liquid fuel firing (switch to gaseous fuels)
CSO2 1 Lfuel Euro / tonne SO2 abated
Specific national data
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
25 Cost of the wet scrubber technical option per tonne of pollutant avoided for liquid fuel firing
CSO2 2 Lfuel Euro / tonne SO2
abated Evaluated by the experts 1,038
(1): Data from Source (1) (2): Data from Source (15)
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
3 Gas turbines in refinery 3.1 General information SNAP CODE: 01 03 04 NFR: 1b Sector activity unit: PJ fuel input SO2 NOx PM VOC NH3 X X X - -
This sector covers emissions from gas turbines in refineries and will be treated in the draft “Combustion”.
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
4 Stationary engines in refinery 4.1 General information SNAP CODE: 01 03 05 NFR: 1b Sector activity unit: PJ fuel input SO2 NOx PM VOC NH3 X X X - -
This sector covers emissions from stationary engines in refineries. This sector will be considered in the draft “Combustion”.
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
5 Fluid Catalytic Cracking Unit 5.1 General information SNAP CODE: 04 01 02 - NFR 4a Sector activity unit: tonne oil throughput SO2 NOx PM VOC NH3
x x x - - Within a refining complex, one of the sources with major potential for atmospheric emissions is the catalytic cracking unit. Emissions from an FCC can be 20-30 % of total refinery SO2 emissions, 15-30 % for NOx and 30-40 % of particulates. [2] This sector is part of the RAINS sector “Other industrial Processes – Petroleum Refining” (code: PR_REF) 5.2 Definition of reference installation/process According statistics from CONCAWE (average capacity of FCC in European refineries = 40,331 bpsd), one Fluid Catalytic Cracking Units (FCC U) with a capacity of 2000 kt/a is proposed as reference installations for the activity of “processes in oil refining”. The flow sheets of a Heavy Oil and Residue Cracker (HORC) – or a Residue Catalytic Cracker (RCC)- are basically the same as for a FCC with the difference that it has a CO-Boiler and catalyst cooler. FCC can be retrofitted to RCC. This technique gives the possibility to upgrade heavier residues than with FCC.
Table 5.1: Reference installations
Reference Code Technique Capacity
[kt/a] feed 01 Fluid Catalytic Cracking Unit (FCCU) 2,000
Data on emission factors from EPA and IPPC reports were exactly the same.[10] [12] 5.3 Emission abatement techniques and costs This chapter considers the emissions abatement techniques for the Fluid Catalytic Cracking Unit in a refinery, the cost of installing and operating them and their performance. The most representative measures considered in the EU-countries are mentioned below. 5.3.1 NOx emissions 5.3.1.1 Abatement measures No primary measures are installed but the SCR and SNCR (in very few plants) technologies may be implemented.
Table 5.2: NOx abatement measures for each reference installation
Description Abatement EF NOx EF NOx
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
efficiency [mg/Nm³] (kg/t feed) [%] Source (1) Source (15) Source (1) Source (15)
Uncontrolled - 800 500 0.48 0.3 Secondary
measures(SCR or SNCR)
80 160 100 0.096 0.06
Remark: An average conversion factor (Fconv) between concentrations of pollutants (in mg/Nm3) and specific mass flows of pollutants (emission factor, in g per t feed) Concentration of pollutant emitted (in mg/Nm3) x Fconv = Specific mass flow of pollutant emitted (in mg/t feed)
Fconv= 600 Nm3/t [1] (2 % O2) [1] 5.3.1.2 Costs The different investments costs proposed by ADEME, which are derived from French refineries [15] are detailed in the Excel sheet established. From these costs, the average investment is 1,300 k€ for SNCR and 5,000 k€ for SCR.
Table 5.3: Investments and Operating costs
Description Investment (k€)
Fixed Operating costs *
Variable Operating costs (€/t)
Source (1) Source (15) (%/a) Source (1) Source (15) SNCR 5,400 1,300 4 0.117 0.0775 SCR 8,000 5,000 4 0.211 0.183
*: The fixed Operating costs only depend on the capacity – or size - of the installation, i.e. on the investment, and they are expressed as a percentage of the plant investment [%/a] Variable Operating costs are defined as the costs depending on the level of production. Parameters for variable operating costs depend on the type of measure (technology) installed. The following tables show the common parameters and prices needed for the calculation of the variable costs. In this case, to determine the operating cost, the SNCR and SCR technologies are considered. The different costs are the following:
� SNCR Electricity cost λe · ce / 103 [k€/t]
• λe: additional electricity demand (= new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh]
λe = 0.112 kWh/t for a consumption of 28 kW [15] (see the Excel sheet for more details) ce = 0.0569 €/kWh (value for France) Ammonia cost λs · cs · efunabated · η / 103 [k€/t]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
• efunabated: unabated emission factor of pollutant [t pollutant/t] • λs: specific sorbents demand (NH3) [t/t pollutant removed] • cs: Ammonia price [€/t] • η: removal efficiency = (1 - efabated/efunabated)
with: λs = λm · λM/η λm: NH3/NOx (mol/mol) ratio for NOx emitted λM: NH3/NOx (mol weight/mol weight) ratio λs = 1.5·(17/46)/0.6 = 0.92 efunabated = 4.8·10-4 t NOx/t according [1] efunabated = 3·10-4 t NOx/t according [15] λs = 0.92 tNH3/t NOx removed [15] cs = 400 €/tNH3 (ammonia pur) η = 60 % Labour cost (λ l · cl ) [k€/t]
• λ l: labour demand [person -year/t] • cl: wages [k€/ person -year]
The number of additional personnel for the SNCR unit is taken as 0.25.
Thus, the annual personnel costs for the SNCR process are:
ACPERS = 0.25 · cl
Thus λ l = (0.25)/ Capacity = 0.25/(2,000,000)
= 1.25·10-7 person-year/t
λ l = 1.25·10-7 person-year/t cl = 37.234 k€/ person-year (value for France)
� SCR Electricity cost λe · ce / 103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh]
λe = 0.68 kWh/t for a consumption of 170 kW [15] (see the Excel sheet for more details) ce = 0.0569 €/kWh (value for France) Ammonia cost λs · cs · efunabated · η / 103 [k€/t]
• efunabated: unabated emission factor of pollutant [t pollutant/t] • λs: specific sorbents demand (NH3) [t/t pollutant removed] • cs: Ammonia price [€/t]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
• η: removal efficiency (= 1 - efabated/efunabated) with: λs = λm · λM/η λm: NH3/NOx (mol/mol) ratio for NOx emitted λM: NH3/NOx (mol weight/mol weight) ratio λs = 1.05·(17/46)/0.9 = 0.43 efunabated = 4.8·10-4 t NOx/t according [1] efunabated = 3·10-4 t NOx/t according [15] λs = 0.43 tNH3/t NOx removed [15] cs = 400 €/tNH3 (ammonia pur) η = 80 % Catalyst cost (λcat · cicat )/ltcat [k€/t]
• λcat: catalyst volume [m3/t] • cicat: unit costs of catalysts [k€/m3] • ltcat: life time of catalyst [a]
In our case, the gas volume flow is around: V = Fconv·Capacity per hour = 600·2,000,000/8,000 = 150,000 Nm³/h Then according some statistics [15], the catalyst volume is around 62 m³. cicat = 15 k€/m3 λcat = 3.1·10-5 m3/t [15] (see the Excel sheet for more details) ltcat = 5 years Labour cost (λ l · cl ) [k€/t]
• λ l: labour demand [person -year/t] • cl: wages [k€/ person -year]
The number of additional personnel for the SCR unit is taken as 0.25.
Thus, the annual personnel costs for the SCR process are:
ACPERS = 0.25 · cl
Thus λ l = (0.25)/ Capacity
= 0.25/(2,000,000)
= 1.25·10-7 person-year/t
λ l = 1.25·10-7 person-year/t cl = 37.234 k€/ person-year (value for France) Table 5.4: Parameters needed to calculate variable Operating costs for SNCR and SCR technologies.
efunabated ηηηη λλλλs
[t/t NOx cs
[€/t] λλλλe
[kWh/t] ce
[€/kWh]λλλλ l
[person-cl
[k€/ λλλλcat
[m³/t] cicat
[k€/m3]ltcat
[a]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
[t NOx/t] removed] year/t] person -year]
SNCR 60 0.92 400 0.112 0.0569 1.25·10-7 37.234 SCR 90 0.43 400 0.68 0.0569 1.25·10-7 37.234 3.1·10-5 15 5
� Conclusion In the petroleum industry for FCC unit, to obtain the cost of the secondary measures, the following shares are taken into account: 95 % of SCR 05 % of SNCR. According to this repartition, the different costs of the NOx secondary measures are the following:
Table 5.5: Investments and Operating costs of the secondary measures
Description Lifetime Investment (k€) Fixed Operating
costs Variable Operating costs
(€/t) (a) Source (1) Source (15) (%/a) Source (1) Source (15)
None - - - - - - Secondary technology 10 7,870 4,815 4 0.206 0.177
Table 5.6: Parameters needed to calculate variable Operating costs for secondary measure
5.3.1.3 Application rate and applicability Respective percentage of reduction measures in 2000 for each reference installation as well as if possible, the percentage of use in 2005, 2010, 2015, 2020 and applicability according to the definition used in the RAINS model. NOx abatement measures
Table 5.7: Application rate and applicability for NOx abatement measures
Description Application rate in 2000
[%]
Application rate in 2005
[%]
Applicability [%]
Application rate in 2010
[%]
Applicability [%]
Application rate in 2015
[%]
Applicability [%]
Application rate in 2020
[%]
Applicability [%]
None Secondary
technologies Dust
application rate
Dust application
rate Dust
application rate
Dust application
rate - For helping to provide the information, use the following methodology. Methodology to calculate the different application rate: The different input parameter to determine the application rate are:
� ENOx: Emission of NOx in a country (t per year) for the different years
� Na: Activity level (t feed per year) for the different years.
Then, the sector situation may be defined by:
efunabated
[t NOx/t] ηηηη λλλλs
[t/t NOx removed]
cs
[€/t] λλλλe
[kWh/t] ce
[€/kWh]
λλλλ l
[person-year/t]
cl
[k€/person-year]
λλλλcat
[m³/t] cicat
[k€/m3]ltcat
[a]
80 0.45 400 0.65 0.0569 1.25·10-7 37.234 3.1·10-5 15 5
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Fs a NOx = (ENOx/Na)
According to this result, it is possible to calculate the different application rates :
FS1NOx: Uncontrolled NOx emission level FS2NOx: NOx emission level implementing the DeNOx secondary measure
The virtual application rate of secondary measures TNOx is obtained by:
TNOx = (Fs a NOx - FS1NOx)/(FS2NOx - FS1NOx)
5.3.2 SOx emissions 5.3.2.1 Abatement measures For SOx abatement techniques, the DeSOx catalyst additive option may be considered as a measure (efficiency of 40 to 50%). The most performing options (Wet scrubbing,…) could be considered in the same category. Table 5.8: Abatement Measure for SOx
Abatement technique Abatement efficiency [%]
SO2 Emission factor (mg/Nm³)
SO2 emission factor (kg/t feed)
Source (1) Source (15) Source (1) Source (15)Uncontrolled 4,000 3,500 0.0024 0.0021
DeSOx catalyst additive option
45 2,200 1,900 0.0013 0.0012
Wet scrubber 90 400 350 0.00024 0.00021 5.3.2.2 Costs The different investments costs proposed by ADEME, which are derived from French refineries [15] are detailed in the Excel sheet established. From these costs, the average investment is 500 k€ for the DeSOx catalyst additive option and 8,000 k€ for Wet scrubber.
Table 5.9: Investments and Operating costs
Description Investment (k€) Lifetime (a)
Fixed Operating
costs *
Variable Operating costs (€/t/a)
Source (1) Source (15) (%/a) Source (1) Source (15) DeSOx catalyst additive option
2,500 500 10 4 1.40 1.25
Wet scrubber 15,000 8,000 10 4 0.939 0.929 *: The fixed Operating costs only depend on the capacity – or size - of the installation, i.e. on the investment, and they are expressed as a percentage of the plant investment [%/a] Variable Operating costs are defined as the costs depending on the level of production. Parameters for variable operating costs depend on the type of measure (technology) installed. The following tables show the common parameters and prices needed for the calculation of the variable costs. DeSOx catalyst additive option
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Labour cost λ l · cl [k€/t]
• λ l: labour demand [person-year/t] • cl: labour cost/wages [k€/ person -year]
The number of additional personnel for the DeSOx additive catalyst unit is taken as 0.25.
Thus, the annual personnel costs for the process are:
ACPERS = 0.25 · cl
Thus λ l = (0.25)/ Capacity = 0.25/(2,000,000)
= 1.25·10-7 person-year/t
λ l = 1.25·10-7 person-year/t cl = 37.234 k€/ person-year (value for France) Additive (zeolithe) cost: λs · cs · efunabated · η / 103 [k€/t]
• efunabated: unabated emission factor of pollutant [t pollutant/t] • λs: specific additive demand [t/t pollutant removed] • cs: additive price [€/t] • η: removal efficiency = (1 - efabated/efunabated)
efunabated = 0.0024 t SOx/t according (1) efunabated = 0.0021 t SOx/t according (15) λs = 0.07 tZeolithe/t SOx removed [15] cs = 18,300 €/tZeolithe η = 45 % Waste disposal cost [k€/t]
• Disposal cost = 152 [€/t of additive] Table 5.10: Parameters needed to calculate variable Operating costs for DeSOx catalyst additive option
efunabated
[t SOx/t] ηηηη λλλλs
[t/t SO2 removed]
cs
[€/t] λλλλ l
[person -year/t]
cl
[k€/ person -year]
Disposal cost [€/t of additive]
DeSOx catalyst additive option
0.0021/ 0.0024 45 0.07 18,300 1.25·10-7 37.234 152
Wet scrubber : Electricity cost λe · ce / 103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh] λe = 2.13 kWh/t for a consumption of 534 kW [15]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
ce = 0.0569 €/kWh (value for France) Limestone cost: λs · cs · efunabated · η / 103 [k€/t]
• efunabated: unabated emission factor of pollutant [t pollutant/t] • λs: specific sorbents demand (e.g. NH3) [ton/t pollutant removed] • cs: sorbents price [€/t] • η: removal efficiency = (1 - efabated/efunabated)
with: λs = λm · λM/η λm: Limestone/SOx (mol/mol) ratio for SOx emitted λM: Limestone/SOx (mol weight/mol weight) ratio λs = 1.05·(100/34) = 1.6 efunabated = 0.0024 t SOx/t according (1) efunabated = 0.0021 t SOx/t according (15) λs = 2.34 tLimestone/tSOx removed [15] cs = 20 €/tSorbent (ammonia pur) η = 90 %
Waste disposal cost
� Dc: Disposal cost = 152 [€/t of waste] � Wa: Amount of waste produced = 1,56 [t/ tonne of limestone]
Labour cost λ l · cl [k€/t]
• λ l: labour demand [man-year/t] • cl: labour cost/wages [k€/man-year]
The number of additional personnel for the wet scrubber unit is taken as 0.5.
Thus, the annual personnel costs for the process are:
ACPERS = 0.5 · cl
Thus λ l = (0.5)/ Capacity
= 0.55/(2,000,000)
= 2.5·10-7 person-year/t
λ l = 2.5·10-7 person-year/t cl = 37.234 k€/ person-year (value for France) Table 5.11: Parameters needed to calculate variable Operating costs for Wet scrubbber
efunabated
[t SOx/t] ηηηη
λλλλs
[t/t SOx removed]
cs
[€/t] λλλλe
[kWh/t] ce
[€/kWh]
λλλλ l
[person-year/t]
cl
[k€/ person-year]
Dc [€/t of waste]
Wa [t/ tonne of limestone]
Wet scrubber
0.0021 0.0024 90 1.64 20 2.13 0.0569 2.5·10-7 37.234 152 1,56
5.3.2.3 Application rate and applicability
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Respective percentage of reduction measures in 2000 for each reference installation as well as if possible, the percentage of use in 2005, 2010, 2015, 2020 and applicability according to the definition used in the RAINS model.
Table 5.12: Application rate and applicability for SOx abatement measures
Description Application rate in 2000
[%]
Application rate in 2005
[%]
Applicability [%]
Application rate in 2010
[%]
Applicability [%]
Application rate in 2015
[%]
Applicability [%]
Application rate in 2020
[%]
Applicability [%]
None A DeSOx catalyst additive option
B
Wet scrubber C
- To support provision of this information, you are invited to use the following methodology: Methodology to calculate the different application rates: In a refinery, either the DeSOx catalyst additive option or the wet scrubber can be installed, but nor simultaneously.
Thus A + B + C = 1 (first equation with 3 unknown parameters)
The different input parameters to determine the sector situation are:
� ESO2: Emission of SO2 in a country (t per year) for the different years
� Na: Activity level (t of feed per year) for the different years
Then, the sector situation may be defined by:
Fs a SO2 = (E SO2/Na)
Using this result, it is then possible to calculate the different application rate:
FS1 SO2: Uncontrolled SO2 emission level FS2 SO2: SO2 emission level after implementing the DeSOx catalyst additive option FS3 SO2: SO2 emission level after implementing the Wet scrubber
Fs a SO2 = A·FS1 SO2+B·FS2 SO2+C·FS3 SO2 (second equation)
But a third equation is needed to solve the system. The only solution is to give the application rate for one technique and then the other could be easily calculated.
Consequently, the different input parameters to determine the application rate are:
� ESO2: Emission of SO2 in a country (t per year) for the different years
� Na: Activity level (t of feed per year) for the different years
� Application rate of one abatement technique
5.3.3 Dust emissions 5.3.3.1 Abatement techniques
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Concerning dust, different configurations of cyclones may be used. The performance range may be greatly variable but as average an efficiency of 80% could be taken. The second option is an end of pipe technology like bag filters and ESP with much better performances. This option could be named “deduster”.
Table 5.13: Abatement Measure for dust
Description PM abatement efficiency [%]
PM emission factor
(mg/Nm³)
PM emission factor
(kg/t feed) None (two stage cyclone) - 600 0.36
Higher stage cyclones 80 120 0.072 Deduster(EP and bag filter) 97.5 15 0.009
5.3.3.2 Costs
Table 5.14: Investments and Operating costs
Description Efficiency
(%) Lifetime
(a) Investment
(k€) Fixed Operating
costs * (%/a)
Variable Operating costs
(k€/t/a) None - - - - -
Higher stage cyclones
80 10 2,000 4 0.039
Deduster(EP and bag filter)
95 10 4 See table 5.15
*: The fixed Operating costs only depend on the capacity – or size - of the installation, i.e. on the investment, and they are expressed as a percentage of the plant investment [%/a] Variable Operating costs are defined as the costs depending on the level of production. Parameters for variable operating costs depend on the type of measure (technology) installed. The following tables show the common parameters and prices needed for the calculation of the variable costs.
� Higher stage cyclone Electricity cost λe · ce / 103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh] λe = 0.6 kWh/t for a consumption of 150 kW ce = 0.0569 €/kWh (value for France) Labour cost λ l · cl [k€/t]
• λ l: labour demand [person-year/t] • cl: wages [k€/person-year]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
λ l = 0 person-year/t cl = 37.234 k€/ person-year (value for France) Dust disposal cost λd · cd · efunabated · η / 103 [k€/t]
• efunabated: unabated emission factor of pollutant [t pollutant/t] • λd: demand for dust disposal [t/ t pollutant removed] • cd: specific dust disposal cost [€/t] • η: removal efficiency (= 1 - efabated/efunabated)
For the considered technique and efficiency, there is no dust disposal. λd = 0 t/ t TSP removed
� ESP Electricity cost λe · ce / 103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh] λe = 0.6 kWh/t for a consumption of 150 kW ce = 0.0569 €/kWh (value for France) Labour cost λ l · cl [k€/t]
• λ l: labour demand [person-year/t] • cl: wages [k€/person-year]
λ l = 3.75·10-7 person-year/t for a number of additional personnel of 0.75 person-year cl = 37.234 k€/ person-year (value for France) Dust disposal cost λd · cd · efunabated · η / 103 [k€/t]
• efunabated: unabated emission factor of pollutant [t pollutant/t] • λd: demand for dust disposal [t/ t pollutant removed] • cd: specific dust disposal cost [€/t] • η: removal efficiency (= 1 - efabated/efunabated)
For the considered technique and efficiency, there is no dust disposal. λd = 0 t/ t TSP removed
� Fabric filter Electricity cost λe · ce / 103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
• ce: electricity price [€/kWh] λe = 2.2 kWh/t for a consumption of 550 kW ce = 0.0569 €/kWh (value for France) Labour cost λ l · cl [k€/t]
• λ l: labour demand [person-year/t] • cl: wages [k€/person-year]
λ l = 3.75·10-7 person-year/t for a number of additional personnel of 0.75 person-year cl = 37.234 k€/ person-year (value for France) Dust disposal cost λd · cd · efunabated · η / 103 [k€/t]
• efunabated: unabated emission factor of pollutant [t pollutant/t] • λd: demand for dust disposal [t/ t pollutant removed] • cd: specific dust disposal cost [€/t] • η: removal efficiency (= 1 - efabated/efunabated)
For the considered technique and efficiency, there is no dust disposal. λd = 0 t/ t TSP removed Table 5.15: Parameters needed to calculate variable Operating costs for primary deduster
efunabated
[t dust/t] ηηηη λλλλd
[t/t dust removed]
λλλλe
[kWh/t] ce
[€/kWh]
λλλλ l
[person-year/t]
cl [€/person-
year] Cyclones 3.6·10-4 80 0 0.6 0.0569 1.25·10-7 37,234
Electrofilter 3.6·10-4 95 0 0.6 0.0569 3.75·10-7 37,234 Bag filter 3.6·10-4 95 0 2.2 0.0569 3.75·10-7 37,234
� Conclusion In the petroleum industry for FFC unit, to obtain the cost of the deduster technology which comprises ESP and fabric filter, the following shares are taken into account: 50 % of ESP 50 % of fabric filter. According to this repartition, the different costs of the deduster measure are the following:
Table 5.16: Investments and Operating costs of the deduster
Description Lifetime (a)
Investment (k€)
Fixed Operating costs (%/a)
Variable Operating costs (€/t)
None - - - - Deduster 30 3,500 4 0.094
Table 5.17: Parameters needed to calculate variable Operating costs for secondary measure efunabated [t dust/t] ηηηη λλλλd
[t/t dust removed] λλλλe
[kWh/t] ce
[€/kWh]λλλλ l
[person-year/t] cl
[€/person-year]3.6·10-4 95 0 1.4 0.0569 3.75·10-7 37,234
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
5.3.3.3 Application rate and applicability Respective percentage of reduction measures in 2000 for each reference installation as well as if possible, the percentage of use in 2005, 2010, 2015, 2020 and applicability according to the definition used in the RAINS model.
Table 5.18: Application rate and applicability for Dust abatement measures
Description Application rate in 2000
[%]
Application rate in 2005
[%]
Applicability [%]
Application rate in 2010
[%]
Applicability [%]
Application rate in 2015
[%]
Applicability [%]
Application rate in 2020
[%]
Applicability [%]
None Cyclones 100 100 100 100 Deduster 100 100 100 100
- To support provision of this information, you are invited to use the following methodology: Methodology to calculate the different application rates: The different input parameters to determine the application rates are:
� Edust: Emission of Dust in a country (t per year) for the different years
� Na: Activity level (t of feed per year) for the different years
Then, the sector situation may be defined by:
Fs a Dust = (E Dust /Na)
Using this result, it is then possible to calculate the different application rate:
FS1 Dust: Uncontrolled Dust emission level FS2 Dust: Dust emission level implementing the Cyclone FS3 Dust: Dust emission level implementing the Deduster
� If FS1 Dust <Fs a Dust < FS2 Dust, it can be considered that some cyclones may still be implemented to a given percentage of the production capacity.
The virtual application rate of primary measures T1, Dust is obtained by:
T1, Dust = (Fs a Dust - FS1 Dust)/(FS2 Dust - FS1 Dust)
� If Fs a Dust <FS2 Dust it may be considered that cyclones have already been implemented. In this case, it can be considered that the application rate concerning cyclones is 100%.
The virtual application rate of deduster T2, Dust is obtained by:
T2 Dust = (Fs a Dust - FS2 Dust) / (FS3 Dust - FS2 Dust)
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Petroleum industry: Fluid Catalytic Cracking Unit Summary list of parameters and data
Parameter Annotation Unit Type of data Current proposal 1 Activity level 2000, 2005, 2010, 2015
and 2020 Na Tonnes per year Input -
2 SOx (as SO2) 2000, 2005, 2010, 2015 and 2020
ESO2 Tonnes per year Input -
3 NOx (as NO2) 2000, 2005, 2010, 2015 and 2020
ENOx Tonnes per year Input -
4 Dust 2000, 2005, 2010, 2015 and 2020 EDust Tonnes per year Input - 5 Application rate of one DeSOx option AR % Input - 6 Conversion factor between
concentration [mg/Nm³] and specific mass flow [kg/tonne of feed]
Fconv - Fixed by the experts
600
7 Uncontrolled dust emission level FS1 Dust mg/Nm³
kg / tonne of feed
Fixed by the experts
600
0.36 8 Emission level after Higher stage
cyclone FS2 Dust mg/Nm³
kg / tonne of feed
Fixed by the experts
120
0.072 9 Emission level after deduster FS3 Dust mg/Nm³
kg / tonne of feed
Fixed by the experts
15
0.009 10 Cost of the cyclone option per tonne of
pollutant avoided CDust 1 Euro / tonne of dust
avoided Evaluated by
the experts 702
11 Cost of the deduster option per tonne of pollutant avoided
CDust 2 Euro / tonne of dust abated
Evaluated by the experts
6,000
12 Uncontrolled NOx emission level FS1 NOx mg/Nm³
kg / tonne of feed
Fixed by the experts
800 (1) / 500 (2)
0.48 (1) / 0,3 (2) 13 NOx emission level implementing the
DeNOx technical option
FS2 NOx mg/Nm³
kg / tonne of feed
Fixed by the experts
160 (1) / 100 (2)
0.096 (1) / 0,06 (2) 14 Cost of the DeNOx technical option
(PM) per tonne of pollutant avoided CNOx
Euro / tonne NOx abated
Evaluated by the experts
2210 (1)
2130 (2) 15 Uncontrolled SO2 emission level FS1 SO2 mg/Nm³
g / tonne of feed
Fixed by the experts
4,000 (1) /3,500(2)
2.4 (1) / 2.1 (2) 16 SO2 emission level implementing the
DeSOx catalyst additive option
FS2 SO2 mg/Nm³
g / tonne of feed
Fixed by the experts
2,200 (1) /1,900(2)
1.32 (1) / 1.2 (2) 17 SO2 emission level implementing the
wet scrubber option FS3 SO2 mg/Nm³
g / tonne of feed
Fixed by the experts
400 (1) /350(2)
0.24 (1) / 0.21 (2) 18 Cost of the DeSOx catalyst additive
option per tonne of pollutant avoided C1 SO2
Euro / tonne SO2 abated
Evaluated by the experts
1,485 (1)
1,366 (2) 19 Cost of the wet scrubber option per
tonne of pollutant avoided C2 SO2
Euro / tonne SO2 abated
Evaluated by the experts
1,000 (1)
0,837 (2) (1): Data from Source (1) (2): Data from Source (15)
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
6 Sulphur recovery plants 6.1 General information SNAP CODE 04 01 03 - NFR 4a Sector activity unit: tonne of sulphur produced SO2 NOx PM VOC NH3
x - - - - This sector covers emissions from sulphur recovery plants (Claus plants) in refineries. 6.2 Definition of reference installation/process Sulphur recovery refers to the conversion of hydrogen sulphide (H2S) to elemental sulphur. Hydrogen sulphide is a byproduct of processing natural gas and refining high-sulphur crude oils. In the widely used multistage Claus sulphur recovery process, a portion of the H2S in the feed gas is oxidized to SO2 and water in a reaction furnace with air or enriched oxygen. After quenching the hot gases to generate steam, the cooled gases are passed through a sulphur condenser to recover liquid sulphur and the gases are reheated. The remaining non-combusted fraction of the feed gas H2S reacts with SO2 in catalytic converters to form elementar sulphur, water and heat. The number of catalytic stages depends on the level of conversion desired. [1]
Table 6.1: Efficiencies of the Claus process [1]
Number of Claus reactors Efficiency (%H2S converted) 1 90 2 94-96 3 97-98
The tailgas, containing H2S, sulphur vapour and traces of other sulphur compounds formed in the combustion section, escapes with the inert gases from the tail end of the plant. Thus, it is frequently necessary to follow the Claus unit with a tailgas cleanup unit to achieve higher recovery. Tailgas from a Claus sulphur recovery unit contains a variety of pollutants from direct process oxidation reactions including SO2 and unreacted H2S, other furnace side reaction products such as reduced sulphur compounds and small quantities of CO and VOC. [1]
Table 6.2: Reference installation
Reference Code Technique Plant size range [t /a]
01 Standard Claus Unit (one thermal and two catalytic steps) 33,333
6.3 Emission abatement techniques and costs
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
This chapter considers the emissions control techniques for the process SRU in a refinery, the cost of installing and operating them and their performance. Only the SO2 abatement techniques are considered. 6.3.1 Abatement measures The most common process is the Claus unit, which enables a recovery rate of sulphur amounting to 95-97%. To bring sulphur recovery yield to 99% or more, a Claus Tail Gas Treating Process can be added to a Claus unit. Indeed gases through Claus plants still contain substantially sulphur compounds. There are different types of tail gas treatment units [13] :
� SULFREEN This process is based on the Claus reaction. Here the sulphur produced is adsorbed on an active alumina based catalyst.
� CLAUSPOL This process is based on the Claus reaction. The reaction takes place in a column with packed beds, with the gas entering from the bottom of the column while a solvent with catalyst is distributed in the top of the column.
� SCOT process The Claus tail gas is selectively hydrogenated to H2S, which is separated from the gas stream in an amine absorber.
� SUPERCLAUS The tail gas is led through a reactor with a selective oxidation catalyst, which converts H2S with excess oxygen to sulphur.[1]
� Others techniques Some others techniques exist also. All these techniques will be regrouped into three categories with different abatement efficiencies as described in the following table 6.3. Table 6.3: SO2 abatement techniques
Abatement technique
Techniques Abatement efficiency
[%]
Emission factor (kg/t sulphur)
Uncontrolled 96 80 Category 1 SuperClaus,.. 99 20 Category 2 Clauspol,
Sulfreen,.. 99.5 10
Category 3 SCOT,.. 99.9 2 6.3.2 Costs
Table 6.4: Investments and variable Operating costs
Measure Code Description
Removal efficiency
Investment (k€)
Fixed Operating
costs
Variable Operating
costs
(%) Source (1)
Sources (16, 17) (%/a) (k€/t)
00 None - - - - - 01 Category 1 99 3,000 2,000 4 3.86
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
02 Category 2 99.5 10,000 5,000 4 2.83 03 Category 3 99.9 15,000 10,000 4 5.11
Source: BAT to reduce emissions from refineries, May 1999. Variable Operating costs [16, 17]: To determine the different costs, according the different costs found in [16], an equivalent quantity of utilities has been taken into account.
� Category 1 Catalyst replacement cost (λcat · cicat) / ltcat [k€/t]
• λcat: catalyst volume [m3/t] • cicat: unit costs of catalysts [k€/m3] • ltcat: life time of catalyst [a]
cicat = 15 k€/m3 λcat = 9·10-5 m3/t for an equivalent catalyst volume of 3 m³ ltcat = 1 year Labour cost λ l · cl [€/t]
• λ l: labour demand [person-year/t] • cl: labour cost/wages [k€/person-year] •
λ l = 7.5·10-6 person-year/t for a number of additional personnel of 0.25 per year cl = 37.234 k€/ person-year (value for France) Electricity cost λe · ce /103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh] λe = 39,1 kWh/t for a consumption of 163 kW ce = 0.0569 €/kWh (value for France)
� Category 2 Catalyst replacement cost (λcat · cicat) / ltcat [k€/t]
• λcat: catalyst volume [m3/t] • cicat: unit costs of catalysts [k€/m3] • ltcat: life time of catalyst [a]
cicat = 15 k€/m3 λcat = 6·10-5 m3/t for an equivalent catalyst volume of 2 m³ ltcat = 1 year
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Labour cost λ l · cl [€/t] • λ l: labour demand [person-year/t] • cl: labour cost/wages [k€/person-year] •
λ l = 1.5·10-5 person-year/t for a number of additional personnel of 0.5 per year cl = 37.234 k€/ person-year (value for France) Electricity cost λe · ce /103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh] λe = 24 kWh/t for a consumption of 100 kW ce = 0.0569 €/kWh (value for France)
� Category 3 Catalyst replacement cost (λcat · cicat) / ltcat [k€/t]
• λcat: catalyst volume [m3/t] • cicat: unit costs of catalysts [k€/m3] • ltcat: life time of catalyst [a]
cicat = 15 k€/m3 λcat = 3·10-5 m3/t for a catalyst volume of 1 m³ ltcat = 1 year Labour cost λ l · cl [€/t]
• λ l: labour demand [person-year/t] • cl: labour cost/wages [k€/person-year] •
λ l = 1.5·10-5 person-year/t for a number of additional personnel of 0.5 per year cl = 37.234 k€/ person-year (value for France) Electricity cost λe · ce /103 [k€/t]
• λe: additional electricity demand (=new total consumption – old total consumption) [kWh/t]
• ce: electricity price [€/kWh] λe = 72 kWh/t for a consumption of 300 kW ce = 0.0569 €/kWh (value for France) Table 6.5: Parameters needed to calculate variable Operating costs
efunabated
[t SOx/t] ηηηη λλλλe
[kWh/t]ce
[€/kWh]
λλλλ l
[person-year/t]
cl
[k€/ person
λλλλcat
[m³/t] cicat
[k€/m3] ltcat
[a]
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
-year] Category
1 0.08 99 11,7 0.0569 7,5·10-6 37.234 9·10-5 15 1
Category 2 0.08 99.5 24 0.0569 1.5·10-5 37.234 6·10-5 15 1
Category 3 0.08 99.9 72 0.0569 1.5·10-5 37.234 3·10-5 15 1
6.3.3 Application rate and applicability Respective percentage of reduction measures in 2000 for each reference installation as well as if possible, the percentage of use in 2005, 2010, 2015, 2020 and applicability according to the definition used in the RAINS model.
Table 6.6: Application rate and applicability for SOx abatement measures
Description Application rate in 2000
[%]
Application rate in 2005
[%]
Applicability [%]
Application rate in 2010
[%]
Applicability [%]
Application rate in 2015
[%]
Applicability [%]
Application rate in 2020
[%]
Applicability [%]
None A Category 1 B Category 2 C Category 3 D
- To support provision of this information, you are invited to use the following methodology: Methodology to calculate the different application rates: In a refinery, all the different abatement measures are not installed together.
Thus A + B + C + D = 1 (first equation with 4 unknown parameters)
The different input parameters to determine the sector situation are:
� ESO2: Emission of SO2 in a country (t per year) for the different years
� Na: Activity level (t of sulphur produced per year) for the different years
Then, the sector situation may be defined by:
Fs a SO2 = (E SO2/Na)
Using this result, it is then possible to calculate the different application rate:
FS1 SO2: Uncontrolled SO2 emission level FS2 SO2: SO2 emission level after implementing the category 1 option FS3 SO2: SO2 emission level after implementing the category 2 option FS4 SO2: SO2 emission level after implementing the category 3 option
Fs a SO2 = A·FS1 SO2+B·FS2 SO2+C·FS3 SO2 +D·FS4 SO2 (second equation)
But two equations are needed to solve the system. The only solution is to give the application rate for two technical options and then the other could be easily calculated.
Consequently, the different input parameters to determine the application rate are:
� ESO2: Emission of SO2 in a country (t per year) for the different years
� Na: Activity level (t of sulphur produced per year) for the different years
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
Petroleum industry: Sulphur Recovery Unit Summary list of parameters and data
Parameter Annotation Unit Type of data Current proposal 1 Activity level 2000, 2005, 2010, 2015
and 2020 Na Tonnes of sulphur
produced by the Claus unit per year
Input -
2 SOx (as SO2) 2000, 2005, 2010, 2015 and 2020
ESO2 Tonnes per year Input -
3 Application rate of one DeSOx option A,B, C or D % Input - 4 Application rate of a second DeSOx
option A,B, C or D % Input -
6 Uncontrolled SOx (as SO2) emission level
FS1 SO2 kg / tonne of sulphur Fixed by the experts
80
7 SOx(as SO2) emission level implementing the DeSOx category 1 option
FS2 SO2 kg / tonne of sulphur Fixed by the
experts
20
8 SOx (as SO2) emission level implementing the DeSOx category 2 option
FS3 SO2 kg / tonne of sulphur Fixed by the
experts
10
9 SOx (as SO2) emission level implementing the DeSOx category 3 option
FS4 SO2 kg / tonne of sulphur Fixed by the
experts
2
10 Cost of the DeSOx category 1 option per tonne of pollutant avoided
C1 SO2 Euro / tonne SO2
abated
Evaluated by the experts
309 (1)
228 (2) 11 Cost of the DeSOx category 2 option
per tonne of pollutant avoided C2 SO2
Euro / tonne SO2 abated
Evaluated by the experts
740 (1)
390 (2) 12 Cost of the DeSOx category 3 option
per tonne of pollutant avoided C3 SO2 Euro / tonne SO2
abated Evaluated by
the experts
955 (1)
693 (2) (1): Data from Source (1) (2): Data from Source (15)
References [1] Best available techniques to reduce emissions from refineries, CONCAWE, Brussels,
1999. [2] Reference document on best available techniques for mineral oil and gas refineries,
IPPC, Sevilla, 2001. [3] Sulphur dioxide emissions from oil refineries and combustion of oil products in
Western Europe and Hungary (1995). Report 3/98. Brussels, CONCAWE, 1998. [4] Technical background document for the actualization and assessment of UN/ECE
protocols related to the abatement of the transboundary Transport of nitrogen oxides from stationary sources, DFIU, 1999.
[5] HOLTMANN, T.; RENTZ, O.: Development of a Methodology and a Computer
Model for Forecasting Atmospheric Atmospheric Emissions From relevant Mobile and Stationary Sources, Volume 3, Part 1, November 1995.
[6] BUWAL: Emissionsfaktoren für stationäre Quellen, Bundesamt für Umwelt, Wald
und Landschaft, 1995.
Petroleum refineries for SO2, NOx and TSP
Draft Background Document 25/02/2004
[7] Description of some refinery installations in Germany [8] RADLER, M.: „1998 worldwide refining survey“, in Oil and Gas Journal Special
(Dec. 21, 1998, p. 49 92) [9] STELL, J.: „Worldwide construction update“, in Oil and Gas Journal (2000 vol. 44,
p.56 – 93) [10] Stand der Technik bei Raffinerien im Hinblick auf die IPPC Richtlinie,
Umweltbundesamt, Austria, 2000. [11] Air pollution Technology Fact sheet on Venturi Scrubbing, EPA. [12] Compilation of Air Pollutant Emission Factors, AP-42, Fifth Edition, Volume I:
Stationary Point and Area Sources, EPA, 1999. [13] Kemp, R.; Sorrell, S.: Technology and Environmental policy, The implementation and
technological impact of the air framework directive (84/360/EEC) in England and Wales, France and Italy.
[14] ER (1996) Dutch Emission Inventory System, TNO-MEP, Apeldoorn, the Netherlands
[15] Etude de faisabilité pour la réduction de NOx de deux raffineries françaises réalisée par le Gaz Intégral pour le compte de l’Union Française de l’Industrie du Pétrole, le Ministère de l’Ecologie et Du Développement Durable et l’ADEME
[16] Claus Tail gas treatments technical & economical evaluations study, elaborated by
Technip for ADEME
[17] Les techniques de désulfuration des procédés industriels, ADEME (Référence 2884)
[18] Reference document on best available techniques for mineral oil and gas refineries, IPPC, Sevilla, 2003.
[19] The Swedish charge on nitrogen oxides – Cost effective emission reduction, Swedish Environmental Protection Agency, 2000.